I was recently involved in a major upgrade at a client site. I won’t name the Client but let’s call them a “SA Business”, although after the success of this upgrade I would sure want to be named.
Around 8 months ago we started to have weekly meetings, and identified that their current version of Sage Accpac, 5.4, needed to be upgraded. This alone would have been fine, however Accpac had interfaces to Sales Logix and the link was critical to business operations. With 80+ users affected by this upgrade, a seamless implementation with low levels of interruption was required.
I had the benefit of being a participant in this upgrade, and it was great to see how well things run when they are planned well. Usually the client relies on us to handle all aspects of the upgrade, but we are often not privy to the big picture and all the IT Requirements. I could plan anything Sage Accpac related and protect the client from disruptions but I could not control the environment or Sales Logix upgrade. Being part of the project team meant the product specialists had all this under control and I could concentrate on what I was good at.
In my eyes these are the things the client did right
- Took the project seriously and committed time, resources and money to have it done right
- Took ownership of the upgrade and had key members from all areas involved
- Hired a project manager to keep the project on track away from conflicting deadlines
- Implemented a test server and performed a “test conversion” in real time to ensure timings
- Had test plans and ensured users completed the tests and signed off on them
Along the way the project grew bigger with new IT Infrastructure being rolled into the mix. The team just rolled this into the planning and everything stayed on track. This would have normally disrupted the normal swing of an upgrade, but these guys took it in their stride.
The “test conversion” to ensure real timings was my saviour, I would have caused this project to fail if it had be a live conversion. I completely underestimated the size of the data and sheer time to move it around. The clients data had to be upgraded from Accpac 5.4, to Accpac 5.6, to 6.0 so this meant a lot of dumps and loads. When we performed the test conversion all was going well, until I realised that if I kept on the same path there was physically no way the data could be converted over night, there was simply not enough hours in the day. So back to the Project team I went with what I considered an epic fail, but this was what the testing was about. We all kicked into gear and came up with a new plan, implemented an interim server on site (granted not every site has these resources) and reduced the speed of the upgrade by around 10 hours.
The night before go live I sat back in my pajamas (remote access is a godsend), converted the data and advised the client that within 3 hours we were ready. Go Live day, was calm and relaxed, everything had been planned, we had a roll back position and everyone knew their jobs. We all ate yummy cakes and smoothly watched the hard work that we had put in come all together.
My advice to any client undertaking even what they consider a minor upgrade is to “get involved and take the time to plan”. Don’t sit back and assume it’s all being done. There are key people on your site who need to be involved, have some meetings, kick some ideas around. If you do this then your upgrade project is ensured to be a resounding success!.
Well Done “SA Business” on what I consider the “Best Upgrade Ever”.